Update

app.py

@@ -4,6 +4,7 @@ import pandas as pd
 from apscheduler.schedulers.background import BackgroundScheduler
 from huggingface_hub import snapshot_download
 from datasets import load_dataset
+import json
 
 from src.about import (
     CITATION_BUTTON_LABEL,
@@ -54,10 +55,9 @@ for bug_id, time in bug_id_to_time.items():
     timeline_ys.append(0)
     timeline_cols.append("Baseline")
 LEADERBOARD_DF = get_leaderboard_df(EVAL_REQUESTS_PATH, COLS, total_issues)
-print(LEADERBOARD_DF["fixed_bug_ids"])
 for row in LEADERBOARD_DF.itertuples():
     model_cnt += 1
-    for fix in row.fixed_bug_ids:
+    for fix in json.loads(row.fixed_bug_ids):
         timeline_xs.append(bug_id_to_time[fix])
         timeline_ys.append(model_cnt)
         timeline_cols.append(row.method_name)

src/leaderboard/read_evals.py

@@ -94,7 +94,7 @@ class EvalResult:
                 (self.build_count - self.build_failure_count) * 100.0 / self.build_count, 1
             ),
             AutoEvalColumn.mttr.name: self.mttr,
-            "fixed_bug_ids": self.fixed_bug_ids,
+            "fixed_bug_ids": json.dumps(self.fixed_bug_ids),
             "method_id": self.method_name + "(" + self.model_name + ")",
         }